Welcome to the open source way. This is our podcast series, SAP's podcast series about the difference that open source can be. And in each episode, we'll talk with experts about open source and why they do it the open source way. I'm your host Carsten Hohage and in this episode I'm going to talk to Claire Dillon about open source and Aspos in the academic world. Hi Claire. Hello everyone. Nice to have you here Claire.
Claire has been working within the open source and inner source ecosystem for a number of years. After many years of working with developer communities in the private sector that is, she has recently transitioned to be a researcher with the University of, I forgot to ask you how to pronounce that, Galway or Galway? Galway. Galway and Lero. Yep, Lero, the Science Foundation, Ireland Research Center for Software, where she's a member of Lero's ASPO team. And Claire also works with a worldwide...
small but growing number of academic OSPOs that there exist. So, and Claire, that means you're located in Ireland and... that popped up in the news recently. Are you still in mourning for... Again, I have to ask if I've...
Ever pronounced him correctly? Shane McGowan? McGowan. Shane McGowan of the Pope. Well, you know, we did have a good wake for him over Christmas and there were lots of opportunities to do sing songs with the fairy tale of New York. So I guess we're missing that opportunity now. Yes, great man, great, great life, great songs.
I've actually never really noticed this fairy tale of New York thing, which must have been their greatest hit of the Pogues, right? Oh, yes. I was more for like, you know, the party-ish acoustic Pogo things. Shane McGowan will be with us every year forevermore.
because of fairy tale for New York in New York. All right, all right. We're not here to talk about music, though. There is plenty of music podcasts out there. So go and listen to them if you want to hear more about the Pokes. This is Open Source. We have actually a first time today. We had SAP colleagues in this podcast. We had people from other companies and organizations. We had representatives of open source.
government initiatives. Your world, Claire, is open source in academia at universities in short. What exactly do you do there? Well, I do wear many hats. So I have worn many hats and I do wear many hats. But more recently, I've been delighted to being part of this community, I suppose, this growing community worldwide of folks that are looking at how to, I suppose...
professionalize more open source within the university context. So as you mentioned, I moved into being a researcher a little while ago. Before that, I'd been working in the private sector and I'd been working in other communities like Inner Source Commons. But now as a researcher, I've been delighted to be part of Liro's OSPA Open Source Program Office because my research focus...
is around open source methods and practices and their use within corporations. So my journey, I suppose, to this area was a kind of a convoluted one. I came initially through InnerSource, which...
as I mentioned, is the, you know, looking at open source practices within corporations because I was interested in these developer practices that were happening in large organizations. And through that, I met a lot of people that were working in open source program offices in the corporate world. And then again, through an evolved...
community because everyone's so close in the open source community I became aware that there was a growing movement of creating open source program offices in public sector and in particular particularly in universities. And so I became more involved working with universities who had set up open source program offices, looking at how they can support open source or operationalize it more consistently within the university. to your institution.
So now, as I mentioned, I'm working with the Liro's OSPO team in Ireland, but I also work with this community of open source program offices in university and research institutions worldwide. It is people paying your money. is the university or is that in academia it's like this fund and that fund or how does that work?
Yes, no. In this instance, my community work around the open source program offices is actually being funded by the Alfred P. Sloan Foundation, which is a philanthropic organization based in the United States. And they're very interested in, I suppose, support.
the open source practices within universities because they believe that that can actually create better quality research outputs. So the Sloan Foundation has funded a number of open source program offices in the United States. But the community I work with...
have those open source program offices and then also a number in Europe too, in particular in Ireland, the Lira 1, as I mentioned, and the Trinity College Dublin, also in Ireland, has had an open source program office for many years as well. So it's a small... growing community I think there's only about
maybe 14 or 15 worldwide known open source program offices in universities and research institutions, but it's a growing trend. And I think that reflects a number of trends in the ecosystem as people are moving. towards open research, open scholarship, open science, as many
I suppose governments start following a more open strategy in terms of software development and innovation, then again, the university sector has responded and is now looking to support that activity. Maybe let's look at where this is moving. in a minute from now. Sure. And, or also at the reason why there is still a relatively small number of hospitals in the academic world, as if I got you right there. In the first place, though,
In how far is and for whom actually at universities is open source relevant or not so relevant? Does it differ from the corporate world or is it basically all the same ideas? Well, that's a really interesting one because I think my big learning over the last while is how different perhaps the university context is than the corporate world. And that's generally speaking because universities and academic institutions have different goals. So in the...
academic world and universities in particular, you can think about open source in the context of education in terms of undergraduates and how they learn about open development practices. You can think about it in the context of how the university might use open source software. for in terms of the tools that they use every day and I suppose most importantly or really more
very relevantly recently, the whole idea of open research has really gained momentum. So as the world has been looking at how do we actually create more reproducible software or reproducible science, should I say, then software forms a big part of that. You know, there's open data, there's open access and open software then is a big part of open research in order to be able to take all the knowledge that's out there and for other people to be able to reproduce it.
test it, build upon it, learn from it. Open software is a great and very important part of that. Okay, but anyway, you did say, I mean, open research or like cooperative research actually in some fields has been around for a while, right? I mean, I'm... The first thing one thinks about is the International Space Station. Yes. CERN or something where international people come together for all kinds of physical, astronomical and so on projects in the two that I mentioned. But you said...
Anyway, that the rise of Ospos is only happening now at universities, now on the steep growing curve, you said. But that means that academia is actually... a bit later than the corporate world it seems why is that Yeah, I think it may again be part of the context of universities. So if you think about the number of researchers who have for years and years, as you point out, you know, often been very involved in the open source ecosystem, but often as an individual. So you often.
have these pockets of open source expertise and activity that are happening in various different academic institutions across the world and great collaborations happening in that space. But often that may be inconsistent within universities or indeed... across different universities in terms of, for example, you know, how that's managed within the university. Is it supported within the university or is that researcher doing it by themselves? How does it align to their IP policy?
Is the local technology transfer office, you know, clued in to how open source works and the trends around commercialization of open source? All of these things can be very inconsistent across various different universities. And I think the idea of an open source program... office within an academic institution has grown in popularity as, I suppose, the sector.
realizes that this is a trend that is growing, becoming more important. And it's important not just to enable it, but to support it and to do it in a consistent way so that...
the folks within the university feel supported in their activities. You know there have been instances in the past where you know academics may think that they're open source sourcing something by just making the source available on GitHub, but not perhaps thinking about the license they may want to add to it, not thinking about the consequences of what license they choose.
And sometimes open source program offices spend a lot of time educating everyone around these considerations and thinking about...
what the end goal is with the research output and with the software that's related to your research, and then thinking about how you want to support that in the university context. That sounds like as if... up to now at universities that basement single-handed developer of or contributor to open source projects still exists up to now and we're now starting to try and get them
under the hood of some larger concept, right? Yeah, but I mean, I'm sure that there are individuals who are involved in the open source ecosystem, but many of the open source projects that come from the academic space are by no means individual basemen.
kind of hobbyist type things. I mean, you have huge projects that are incredibly healthy, incredibly large worldwide. It's just that they may not be supported within the individual institutions. So if you have things like SciPy or, you know, the whole...
scientific Python community. These communities are massive and incredibly professional, but perhaps not being, I suppose, there's a question as to what degree in which universities they're being supported. Now, that's not to say actually that there isn't a lot of supports.
for open source in universities existing today. So there are a lot of activities that happen in some universities in an open source program office that are actually happening in individual universities. They may not just be centralizing that activity and calling it an open source program office. But of course, there are many universities worldwide who are very advanced in their support of open source.
I think the rise of the open source program office allows for individuals involved in that activity to kind of have a flag in the ground to say this is the place in the university where you can come and get information about this and allows for this networking across. university. So within the community I work with, there are a lot of folks who share resources and share concerns and challenges and talk about those sort of things. So I don't want to suggest that
the kind of activities that are often found in open source program offices aren't happening to date in universities. Many universities are doing that. But I think the formalization of those activities into an office is what's new. Okay, so... Then what does one do in this phase for that formalization and how actually, I was just thinking when you were speaking, how do you ensure?
But you're actually supporting these already pretty well managed projects and not interfering with them with additional rules and regulations that they may never have wanted. Well, that's an interesting one, because I think, again, one of my learnings in the academic space is that rules and regulations aren't quite considered in the same way as they may be in the corporate world. So it's nigh on impossible in an academic context to...
you know, make one rule to rule them all in terms of how you approach these things. That may be possible in a corporate context because the goals may be more consistent in a corporation. But if you look at a university, you've got people in the computer science department, you've got people in humanities, you've got people...
in the health department, you've got people in engineering and they're so different. And indeed, their goals around open source may be incredibly different. So it's a very diverse population. Some people are very... technical some people are not very technical and there are a lot more considerations at play because you've got things like you know how you respond to funding grants you've got
local government policies around how and where research outputs should be published. There's the broader context of the fact that researchers are, I suppose, In many respects, they're always looking for citations and references. It's hard to do that with software. So how are they rewarded for contributing software to the academic community at large? So many of these considerations are things that I think...
are important to consider. Certainly, you know, when I think about the new Ospos that have emerged over the last few years. Oftentimes, they're very involved with just figuring out what kind of open source activity is happening at universities, therefore looking at the supports that are necessary. And that may be different per institution, depending on, you know, how advanced the competency around open source.
is within their university. Sometimes it's just about connecting the people who are doing it really well with people who are maybe new to open source within the community. Sometimes it's awareness and advocacy around open source so that a broader set of people understand the opportunities and the challenges. associated with open source and oftentimes a lot of the time in fact it may be also support around this idea of open source licensing giving people more guidance around
how and why to choose a license given your goal around the open source project. So there's a lot of education, there's a lot of guidance. And in what, like you personally or the university you're working with, in what... phase would you say are you? Are you already providing the tools, the guidelines, the information to support projects or are you still in the analysis phase? Do we even have happening in our space of influence? Well, certainly.
In Liro's case, and Liro, of course, is a research institution that actually spans 12 universities within Ireland. So it's a group of folks that are involved in software development. A lot of the activity in Liro's case is around this idea of education and providing guidance.
support to folks that may have I suppose questions around that and doing advocacy around open source. A recent project we're involved in has been funded by Ireland's National Open Research Forum and we're looking at helping all the universities perhaps put some framework or policy in place to help get clarity.
within the Irish University context about how to think about open source or how to think about the goals and what license you may choose and things like that. So that's one of the projects we're involved in. As I mentioned, though, depending on the university, the open source program office may be at different levels. So...
There are some open source program offices who are already working with individual projects to help them grow community who have already been successful in terms of publishing an open source project. There are some universities who are focused on even providing infrastructure to support researchers do open source more efficiently. I mean, in those cases, they might be working very closely with the IT departments. And again, I failed to mention, but it's important to state that...
oftentimes the activities of an open source program office within a university can also be dependent on where that's placed within the university. So in Ireland, for example, in the Trinity, OSPO I mentioned, the OSPO is actually in the technology transfer office. So they're very concerned with actually helping researchers spin out or commercialize some research in that open source context. That's a lot of their competency there. But in other parts of...
the world, the OSPO may be placed in the library and therefore it may be looking at the whole idea of the software as another research output, like papers, like data, and they're thinking about managing that for the university context.
and how that can be done. Or in some cases, it might be in a research institution, in which case... you know, if it's part of a particular discipline like computer science or something like that, then they may be more focused on thinking about open source in that context and therefore getting more specific about particular projects. Remembering my own way back time at university, I'm a geologist by trade actually.
And we had this illegally good equipment with computers back in the 90s, illegally, well, against regulations at least. We shouldn't have had as much compute power at that. in that department without offering, I don't know, a couple of dozen of hours of lessons that would have focused around computers and software. Our head of department was very good at hiding that. Is that still a problem that some people may not even be one to found out with what they do?
I mean, there's also that cultural element within universities. So I think it's been, whereas, you know, researchers, even in the context of open research, do, you know, want to... publish software in order to make their research reproducible. They may not want to do it too early in their actual research project. So there's definitely a trend towards some researchers for perhaps actually, you know, perhaps
keeping their source code closed until they've published their research output and then actually open sourcing it at a later date. Now, you know, I think it's interesting even to consider this idea of software being released as open source as part of a research output. because in those cases, those projects may not have the same goal, for example, as a community open source project because if a researcher has done a piece of research...
released the output, they may have no personal interest in actually growing that community and actually maintaining it for years. They may be moving on to another funded project and have no time to maintain the project that they were working on three to five years ago. So it's a different dynamic because, you know, in the corporate world or in the broader open source community.
You often have individuals who are involved in a project for a very long period of time. And that may not be the case in academia, where if you're releasing a piece of code, it may only be... relevant for a period of time it may not actually have even planned longevity it may just be a piece of software that was only relevant for that particular piece of you know that
particular project work so that gives a different dynamic actually in universities and to your point it can create I suppose you know when you're thinking about collaboration around these projects then you have to be really clear about the expectation and the goal that why you're releasing the software, because otherwise people can get confused about that.
and have expectations of maintenance and engagement that are not going to be fulfilled. I guess if we would want to summarize and compare corporate world and academia, The importance of intellectual property on the corporate side is a matter of the organization. And in the universities, it's a matter of the individual who want to protect their own. And the organization. Yes. Which actually makes it quite, you know, and the funding agencies and actors. Yeah.
So it's a complex. I think it's a bit shifted to the individual who have to protect their own research, their own intellectual property, their own academic career in the end, that is. Oh, it's just far more complex when you think of the actor.
at play because you do have these individuals, you have the teams that work with them, they may have research assistants, there may be collaborations happening across institutions, there may be collaborations happening with the private sector. And in all these cases, this idea of IP and who owns it. Fantastic.
how it's a much more complex scenario. And I remember that the discussions sometimes already start between the researcher and their assistants who owns what. Yes, indeed. Yes, but now that we've boiled this pretty much down to researchers and then the... very beginning you said for undergrads it's more like we learn about open source and the researchers is more like we have an interest in using contributing creating our own let's take it maybe one step further in how far does
the intersection to the corporate world play a role. As in, we've had this episode with this veteran from the University of Waterloo in Ontario, Canada, who already in the 1970s or 80s had these transitions from university grounds into free market spin-offs and hybrids between that and how far is that important?
Yeah, I mean, it's an incredibly important part. And I mean, I referenced before Trinity's open source program office was formed within the technology transfer office. And that was to enable actually what they saw as this trend of some researchers working in the open source world and to support them.
doing spin-outs on open source technology. And they have a couple of examples that they've supported in that respect that it just made sense for those organizations' business model to work in open source because that's how the community they were dealing with worked. And they were concerned with and wanted the wide and rapid distribution of their technology, which is why they chose to open source their code, but then built businesses around that. So it was in that recognition that Trinity...
kind of created their open source program office to support those kind of organizations. And I think, you know, one of the things that universities are very, I suppose, keen to do... is affect these collaborations with the private sector because that can often lead to more grant money or more funding to help the university make research happen. But I think it's really important because actually, as a vehicle...
for translating research into real impact in the ecosystem and in the private sector. I think it's a really potentially good path to consider because if you think about how it might happen otherwise, which is where... You know, a corporation might say, I want a piece of research done. They pay for it.
the researcher goes off and does it. There's a protracted length of time where the lawyers all talk about IP transfer. And then the researcher kind of hands it over in a bunch and says, now I'm off to another project. Like that can sometimes not necessarily be the best way to... embed technology in an organization's own.
Which actually, let me, sorry, interrupt there, because that leads us to the ever unsexy aspects of maintenance and long-term... long-term keeping and everything of projects, is that supported by these spin-offs better? Well, I mean, you have the spin-out scenario where you have the individuals involved often going with the spin-out to actually maintain that project in the future. But even...
Outside that, even if you think about collaborations between academic institutions and individuals in there and private institutions, sometimes if that's done in the open, you have these...
personal connections that are made between the folks that built the code, the researchers, and the folks that are consuming the code or will use it afterwards, which would be the private sector. And that can provide potentially for better collaboration. Also, you know, if you're building an open source project or if you're providing software as...
open source. There's such a much bigger emphasis on things like documentation, making sure people can use it easily. Like all of the best practices around open source project development can actually facilitate much more effective research translation into the private. sector I think or at least in certain circumstances that can be optimal and even when you think about that
in a broader context where you may have multiple organizations and multiple institutions all collaborating on something that's really going to drive forward an industry, that can be hugely powerful. And being able to take away any of the friction that might...
be there where people are having little kind of, I suppose, tensions over who owns the IP. If it's all in the open, that can sometimes make those collaborations happen much more easily. And actually, for me, that's a huge opportunity around this growth in, I suppose. Thinking about growing the open source ecosystem within universities, that's often the goal here. I seem to remember that when we talked before we are recording this today, you had an example.
Something about Internet of Things. Yes. Only recently, I've been speaking with an individual from the European Commission, Max Lemke, who was talking about, he's part of looking at the Internet of Things across Europe. And he was talking about the opportunities around, for example, the software-defined... vehicles and how that's a growing ecosystem.
in open source across Europe and just thinking about those collaboration opportunities at an industry level, both for the corporations, large and small, that may be getting involved in that, but also those academic researchers who may be able to then find a place for their research that actually has immediate commercial impact, that's really where that opportunity is. Rather than working individually, isolated, where you're...
building something theoretical, but it's not having this huge impact on the industry. I think within the context of the broader open source ecosystem, it just allows for much more impactful research to be translated. almost immediately into something that would actually help the entire industry. Yeah, and I think this should actually be hugely attractive also for the researchers at universities because it kind of, when I think back,
I had to make a decision at some point. Do I want to stay in research or do I want to go look for a job outside? And this kind of seems like you can pop your head out a bit. And if you don't like what you see on that other side, you pop it back. Exactly, test the waters. That sounds very neat actually. Now, the next thing I was gonna ask you.
You've mentioned quite a few in the course of this organization. Should we maybe go through them one more time and kind of do a quick characterization of what they do? You mentioned Sloan. You were, I think, paying your salary, you said, right? Or you have grants from. I don't think... Have we mentioned chaos on the way? We haven't yet, but I do think... We have not yet? Yeah. You've mentioned some other ones, Irish specific ones. Let's maybe take some of those off. What does Sloan do?
Well, so I suppose to kind of cover some of the organizations that are at play in this area. So Sloan, Alfred P. Sloan Foundation are, as I said, a philanthropic organization that are very keen to support the creation of these institutions within universities. to support research outputs, to make them better, to make them more effective and productive. They've done great work in that space. But if people are interested in this entire area of open source in academia and learning about...
the context and talking to other people who are working in that context. I suppose there are a few open source communities that are really active in this space that I would really recommend people get involved in and come along to. So one of those is... as you mentioned, the Chaos organization, that they're the community health analytics for open source software. They're a project under the Linux Foundation, and they look at metrics and measurements for healthy.
open source projects across you know the entire open source ecosystem but they have an academic or university working group where it's looking at specifically how do you measure healthy open source within the university context so that's a great place to if you're interested in and metrics and, you know, models of metrics and frameworks and all that sort of thing. They're a great place to actually go and discuss that. There's another organization called SustainOSS and their community is...
It's a place where people are discussing how to sustain the open source ecosystem. What are the opportunities and challenges around that? And thinking about open source in the long haul, what are the systems we need to put in place to sustain open source? They also have an academic working group where they talk about open source in that academic context and thinking.
about the various different players. It's a great place actually to go and find out about all the players and all the organizations that touch on this because, I mean, you know, I've only mentioned a handful here, but there are others where there are crossover interests in terms of open.
So another organization, for example, would be the Research Software Alliance. I mentioned that there was a rise in open research, and that is definitely the case. There's also been a growth of momentum around this idea that as we think about the importance of...
software within the academic context we need to recognize the people who are building the software that help other researchers do their job this idea of that being a discipline in itself is one that's growing momentum and the research software alliance looks at that now as part of that
Of course, open source plays a part. They're not exactly identical goals, but open source, if you're trying to build software that supports globally the research community in doing better research, you want to make sure that's open source because that helps everyone. If I got you correctly, then Sloan and the Research Software Alliance, was it, right? They are the ones that have a certain...
specificity to academia, right? And the other ones are just general open source organizations. Well, in both the case of chaos and sustain, they have working groups that are specific to the academic situation. Yeah. Okay. Yeah.
Okay. And chaos, by the way, is spelled with a double S and has nothing to do for our German listeners with the original German hackers of the chaos group. Gosh, no. No, no, no. By the way, these hackers are now mostly security consultants and not hackers anymore, for all I know at least. Now, as we talked about different organizations, let's turn it around. Is there anything that SAP or companies of the likes of SAP could do to foster open source at universities and other institutions?
Well, yes, indeed. I mean, we were talking about these potentials for industry academic collaborations. And I mean, SAP, I know, do great work in terms of open sourcing projects. You can bet your bottom dollar that you already have contributions coming from the academic community or students or researchers already. But perhaps there's an opportunity to, I suppose, identify those folks and celebrate that sort of thing, but also a potential for corporations to look.
for these opportunities for collaborating with academic institutions to, I suppose, get that idea of taking cutting-edge research from the academic space into projects that are happening today. So there's a huge... opportunity around that, I think. And even bringing that up as a potential pathway to do these kind of collaborations, I think, would be fabulous. So for me, that's the huge opportunity for SAP and others.
I actually know that we're definitely doing a ton of that with the regional universities in Germany. I don't...
know in how far internationally, just also wanted to give you the chance to make that call. Oh, well, I'd be delighted to hear about that. We should be highlighting those cases. But let's come as we're running a... bit out of our regular time here the before last question where do people go to get more information about your work or where do researchers go as a first point of contact when they want to look at open source in their context.
Well, I think for researchers, I mean, a first port to call would be just to find the communities that are active in your space. So if you look at any topic plus open source, you'll probably find a project that's specific to that area. So that'd be the first thing, like go find your tribe of people doing open. source.
within your context. But if you're interested in talking about this idea of open source more generally in the academic space, I hang out in both sustainoss.org, the academic working group there, and in the Chaos University working group.
Both of those organisations hold regular virtual meetings. And physically, I'll be going along to, for example, the Chaos Con in Fostem in just next week, actually, in Brussels. Really looking forward to that. They're the communities, I think, which are having that discussion about...
open source in academia at this kind of higher level and thinking about that more broadly across the whole sector. Yeah, be great to see you there. Or in fact, reach out to me personally. I'm on LinkedIn. It's easy to find me there. Claire, no I, C-L-A-R-E. in LinkedIn. Be delighted to connect with anyone who's interested in this space. All right, you heard her. Just get in contact and we'll of course put all the links under the...
podcast when we publish it. That leads me to the final question. What are the three to four main things, the key takeaways you would want everyone to remember from this episode? Well, I think for me, the three takeaways is that open source is incredibly important in the university context, in education.
but particularly this idea of open research and as a research output, incredibly important. And it's different. It's probably different than the corporate context because sometimes you don't want to have projects lasting forever. So that's different. And I think the real takeaway for...
this is that we all need to be talking more about how to better do it in the academic context in the university context and yeah I'm looking forward to continuing that conversation with everyone who's interested in that area. Okay, then let's hope you get tons of input.
And I thank you very much, Claire, for being our guest today. It was nice to have you here. And thank you all out there for listening to The Open Source Way. If you enjoyed this episode, please share it. Don't miss the next one. We usually publish every... last Wednesday of the month and you'll find us on open SAP and in all those places where you find your podcasts be they proprietary or be they open source podcast clients thanks again
Thanks again, Claire. And bye-bye. Thank you, Kirsten. Bye, everyone.